II. CHEMISTRY 



187 



drazorie-like bodies with o-phcnylenediamine,^* pheiiylhydrazine,^-" '''•' and 

 with various siibstitutccl phcMiylhydraziiics.''- 



The precise nature of the ring in I, the final point to l)e estal)lished, was 

 proved by the well-known methylation techniciues previously developed 

 for the specific purpose of determining the size of the rings in sugars and 

 their derivatives.^*' The neutral 2,3-dimethylascorbic acid (VIII) prepared 



CO-i 

 I 



MeO— C 



II 



MeO— C 



H— CO- 



HO— C— H 

 1 

 CHsOH 



VIII 



CO—, 



I 

 MeO— C 



II 

 MeO— C 



H— CO— I 

 I 

 MeO— C— H 

 I 

 CHjOMe 



IX 



CO- 



I 

 MeO— CO 



MeO— CO 



H— CO— J 



MeO— C— H 

 I 

 CHjOMe 



X 



CONH2 



I 

 CONHj 



CONHj 



H— C— OH 



MeO— C— H 

 I 

 CHjOMe 



XI 



from I by methylation with diazomethane w^as completely methylated with 

 silver oxide and methyl iodide to give the corresponding tetramethyl de- 

 rivative (IX). The close similarity of the selective absorption bands shown 

 by un-ionized ascorbic acid (I) and its di- (VIII) and tetramethyl ether 

 (IX) proved that no alteration in ring structure took place during these 

 methylation experiments. Cleavage of the double bond in IX was effected 

 with ozone to give the neutral mixed oxalic ester (X) . The latter readily 

 reacted with ammonia with the formation of oxamide and the dimethyl- 

 L-threonamide (XI). Isolation of oxamide furnished further proof that a 

 double bond is located between C2 and C3, while the characterization of 

 XI as 3 , 4-dimethyl-L-threonamide by the fact that it gave a positive 

 Weerman test for a-hydroxy amides finally proved that the ring of ascor- 

 bic acid was five-membered and engaged d and C4. The formula (I) as- 

 signed to ascorbic acid w'as thus fully substantiated. ^^ 



A molecular model of ascorbic acid built on the basis of formula I will be 

 seen to be quite flat. This is in agreement with the x-ray-crj^stallographic 

 data furnished by Cox and his associates in the early days of the structural 

 investigations into vitamin C.*' This is of some interest perhaps, because 

 until it was discovered sometime later that dehydro vitamin C was a lactone 

 and not an acid, the x-ray results provided the only evidence which did 



" H. Ohle and H. Erlbach, Ber. 67, 555 (1934). 



" H. Ohle and G. Bockmann, Ber. 67, 1750 (1934). 



" W. N. Haworth, The Constitution of the Sugars, Arnold, London, 1928; E. G. V. 



Percival, Structural Carbohydrate Chemistry. Prentice-IIall, New York, 1950. 

 5' E. G. Cox, Nature 130, 205 (1932); E. G. Cox, and T. H. Goodwin, J. Chem. Soc. 



1936, 769; E. G. Cox and E. L. Hirst, Nature 131, 402 (1933). 



